This proposal continues our studies on class I molecules and how mutation of the genes that encode these molecules affects T cell recognition. One mechanism to explain the multiple base changes seen in H-2Kbm mutant genes is gene conversion. The Q10 gene has the bases found in the mutant H-2Kbm1 gene and has been proposed to be the donor gene for this mutation. We will generate CTL against Qa antigens, including those not normally expressed (such as Q10), and determine the relationship between these antigens and H-2Kbm molecules (and other class I molecules). Interferon has been shown to up-regulate the expression of class I antigens. We will demonstrate that interferon does not up-regulate some Qa antigens. Using domain shuffled Qa/class I antigens, we will determine the region of class I genes that allows for interferon inducibility and determine the role of interferon in augmenting CTL activity. The structure of Qa and TL antigens is very similar to that of the class I molecules, H-2K, D, and L. Nevertheless, the former molecules do not serve as restricting elements for antigen specific CTL. Since domain shuffled Qa and TL antigens are expressed on transfected L cells at relatively high density, we will test the ability of these molecules to function as target molecules for antigen-specific H-2 restricted CTL. In collaborative studies, these Qa and Tla exon shuffled genes will be introduced into mouse ova and the transgenic mice will be tested for their ability to use these molecules as restricting elements for antigen-specific CTL. Site specific mutagenesis of MHC genes allows for a determination of what structures affect CTL recognition. We will determine the fine specificity of anti-H-2Ld CTL clones by testing their activity against several H-2Ld mutants. In collaborative studies, the structure of the T-cell receptor genes will be determined. The role of carbohydrate on MHC molecules in affecting CTL recognition will be determined, as will the affect of removing the entire cytoplasmic portion of the molecule. Finally, the structural components of VSV will be analyzed to determine 1) which molecules encoded by the virus are recognized by CTL and 2) whether mice with differing H-2 alleles respond to different viral components.